[tei-council] a proposal for a change to ODD (copy of ticket I just put in SF)

Laurent Romary laurent.romary at loria.fr
Fri Mar 5 02:23:37 EST 2010

I think you're right and in my talk in Ann Arbour, I started to think  
of mechanisms to simplify the move.
First, instead of introducing a new element to the purpose of  
indicating what one wants to use, we could just invent a new mode, like:
<elementSpec ident=”sexyThing" mode=”use”> (and probably do something  
similar for moduleRef)
Second, we need to simplify the listing of needed elements by  
introducing a simplified version of 'cristals', so that by saying:
<cristalSpec ident=“biblStruct” mode=“use”>, we tell the processor  
that we want biblStruct and all (mandatory?, defined in the same  
module?, classes?, ) elements directly below. We could then in turn  
have a few 'delete' instruction to further reduce the thing.

Le 4 mars 10 à 17:38, James Cummings a écrit :

> Sebastian Rahtz wrote:
>> On 4 Mar 2010, at 15:44, Gabriel Bodard wrote:
>>> This is not really an objection, but I note that I quite like  
>>> being able
>>> to re-run my unchanged ODD against the new version of Roma and see  
>>> the
>>> new element <surplus> added to my schema, because it's in the
>>> modules/classes that I've already said I want.
>>> Is your proposal a replacement for, or a supplement to, the  
>>> existing model?
>> definitely a supplement. existing ODDs work as before
> Yes, I mean I would assume that you can have ODDs written exactly
> as before and saying <moduleRef key="foo"> then you want the
> entire module and whatever is it in the future.  If instead you
> only want certain elements you wouldn't have the moduleRef at
> all, and instead just have lots of elementRefs.  Or am I
> misunderstanding the mechanism?
> -James
> _______________________________________________
> tei-council mailing list
> tei-council at lists.village.Virginia.EDU
> http://lists.village.Virginia.EDU/mailman/listinfo/tei-council

More information about the tei-council mailing list